The language of science

This is a post I've been elaborating in my head for a long time. The recent retraction of a PLoS ONE paper with unusual language that may have been a translation error, and the associated debate, has prompted me to get it written down. It does not directly address the so called  #creatorgate, but it may help think about issues of language and power in science. If science is truly to be an international collaborative enterprise, then we need to become more conscious of what conducting that business in a language that is not the native one of many scientists means.

Languages matter to me. Like my siblings, I was raised fully bilingual. I can read and write fluently in both English and French, and at one point was an in-house translator for the channel-spanning company I worked at. Yet, when it comes to science, I am functionally monolingual: I cannot write about the work I do in French with any confidence that the words I use are correct, and that my explanations are meaningful.
Oddly, it was not always that way. My entire schooling up through highschool was in French, so I went to University knowing all my scientific words and concepts in French, not English. I had to cobble together an English science lexicon quickly, and for the longest time, there were certain things (integration and differentiation in particular) that I would do in French in my head. Yet, after 15 years of doing science exclusively in English, the French has been displaced. At this point, the only thing I do in French is long division, which is a skill I maintain much as one might practice fencing with a rapier. And here is the point: there is no incentive, professionally, for me to learn to do science in French. I already do science in the de facto lingua franca.
When I joined my PhD program, my advisor lamented that previous students had successfully campaigned for the  abolition of the language requirement. At the time I was saddened, because I am always sad when people pass up the opportunity to learn a second language (and because I had just missed out on an easy credit). Now I am irritated. Irritated at the myopa of native English speakers decreeing that language requirements are unnecessary for PhD programs that demand English as Second Language certifications for all non native-English speaking applicants.
Because, when we get down to brass tacks, science today does have a language requirement: speak English. I want you to let that sink in for a moment. If you want to work where the jobs and money are, if you want your work to be cited, you have to speak English. So hegemonic is English's position as the language of science that, in many Universities with global ambitions, one can be hired as a professor there without speaking the local language, as long as one speaks English. Before you tell me that this is a sign of the internationalism of science, let me point out that the converse is not true. Anything but.
Let us take a minute to think what it means that a language that is spoken natively by 6% percent of the population is a sine qua non of doing science. If you've ever tried to master a foreign language to the point of being able to travel in that country, you know it's hard. Imagine doing it to the point of being globally competitive in your field. And let's add that we offer no help: no science societies, or universities, or journals, pay for translation services, or language classes. And, in fact, we scoff and are suspicious of letters by applicants from those countries as not having been written by the applicant (as if English-speaking applicants did not get their application documents heavily edited). We grimace when people have trouble clearly expressing themselves in English, without acknowledging that this is a challenge we will never have to face with anything like the same consequences (jobs, publications) for butchering a talk we might choose to give in Japanese or Mandarin or Hindi or Urdu (which we would only ever do as an outreach exercise anyway). Now, I want you to realise that every single foreign postdoc or grad student or faculty member in your department has put in the work to be good enough in English that she can just about communicate science with you. If you tried to order lunch for her in her native language, how far would you get?
Yet, beyond the obvious dreaded-P-word that is attached to being a native English speaker, and the obvious selection bias for people with ressources that it places on scientists from non English speaking countries, there is a more insidious effect, which I alluded to in the beginning. When certain fields are conducted only in one language, other languages loose out. For example, my mother speaks Alsatian (the Germanic dialect of her region of France). As a dialect, Alsatian is a language that important humanistic scholars (and Goethe's mentors) would have used. But, as French and Hochdeutsch became the languages of French and German nationalism, Alsatian, along with other dialects, was squeezed out of law, administration, science. Recent censuses put current Alsatian usage at about 60 to 70% of the population of Alsace, huge for a regional language in a country whose relationship to regional languages is ambivalent at best. And yet, my mother, who learnt her Alsatian from a woman born in 1870, knows words no one now knows. As the sphere of topics discussed in Alsatian has shrunk over the centuries, all the language associated with abstract or technical constructs has been lost. Yes, many people still speak Alsatian, but only when discussing the most mundane of topics.
As a Franco-British person growing up in England, France's often ham fisted attempts to promote French neologisms against English borrowing were often lampooned. And certainly there is a contemptible side to a former imperial power lashing out against another imperial power that has awkwardly gained ascendancy. Yet there is something equally contemptible in the tendency of Native English speakers to view the hegemony of English as a "natural" process. If we stop to consider the historical forces that have lead to this state of affairs, is this something to which science truly wishes to unequivocally yoke itself?
Because there are other models. At least two international organisations, the UN general assembly and the EU parliament advocate multilingual systems with active translations, because they recognise that equal participation cannot require someone learning a whole new language. And yes, both the UN and the EU translation administrations are hugely expensive. But if, as a publishing company, you're making billion of dollars in profits while maintaining a 40% profit margin, maybe translation services could be part of the added value you offer? And, if you're fighting for the creation of publicly funded open access repositories, and repeatedly tell me (because I've asked this question) that you have the long term digital archiving question all handled, then maybe you can also find money to broaden participation by supporting translation services?
Because, if not, and you're a native English speaking academic talking about how you're broadening access while requiring an ESL certificate from overseas grad students in your department? Then don't conflate the dregs of 200 years of imperialism and naked political power play with internationalism.